Search Console guide · Updated August 2026
Google has finally given site owners a separate view of their visibility inside AI Overviews, AI Mode and generative features in Discover. That is useful progress. It is not, however, a complete AI attribution report.
The new view tells you where your URLs are appearing. It does not yet explain the prompt that triggered the appearance, the exact passage Google used or the revenue that followed. The practical job is to combine this new visibility layer with the Search Console and analytics data you already trust.
What Is the Generative AI Performance Report?
Google launched dedicated Search and Discover Generative AI performance reports in June 2026. The Search report covers appearances in features such as AI Overviews and AI Mode; the Discover report covers generative experiences in Discover.
The data is not entirely new. Google says these impressions remain included in the overall Performance report. The dedicated view isolates them, making it far easier to see which pages have visibility specifically inside generative features.
What the Report Shows—and What Each Dimension Means
| Dimension | What Google reports | How an ecommerce team can use it |
|---|---|---|
| Impressions | How often a URL from your site appeared in a generative feature. | Track visibility trends and compare product, collection, guide and policy-page performance. |
| Pages | The URLs that appeared. | Find the pages Google repeatedly selects and pages that lost visibility after an edit or migration. |
| Countries | Visibility by country. | Check whether the right regional store, currency and shipping information are surfacing. |
| Devices | The device used when the result appeared; available for Search. | Spot mobile-heavy discovery and prioritize mobile UX on the landing pages being exposed. |
| Dates | Hourly, daily, weekly and monthly views. | Separate short spikes from durable growth and annotate releases, promotions and content changes. |
There is a crucial language choice here: Google calls these impressions, not recommendations or endorsements. A URL can be visible while receiving no click. It can also appear as one supporting source among several. That is why raw totals should never be presented as “AI traffic.”
How to Use the Report Step by Step
Four Analysis Views Worth Saving
1. AI visibility winners and losers
Sort page-level impression change from highest to lowest. Add the page type, last substantial edit and ordinary Search clicks. This quickly shows whether growth came from genuinely useful commercial content or from a one-off informational mention.
2. Visibility without traffic
Flag pages with rising AI impressions but flat organic sessions. Do not call them failures automatically. Check whether branded search, direct visits, product-page entrances and assisted conversions moved too. Our guide to content that performs in both traditional results and AI Overviews explains how to preserve the next-click value.
3. Regional mismatch
Compare impression countries with the markets each page serves. If a US URL is appearing heavily in the UK, improve currency, delivery, returns and hreflang signals before celebrating the reach.
4. Content-type mix
Calculate the share of impressions earned by product pages, collections and guides. If editorial content owns the visibility but commercial pages are absent, strengthen contextual links and make product evidence easier to find. A structured Shopify SEO audit is a good starting point.
How to Measure Impact Without Overclaiming
The safest reporting model separates three layers:
Do not calculate an “AI CTR” by dividing standard Search clicks by AI impressions. Those numbers do not describe the same isolated surface. Until Google adds click data to the dedicated report, report them as separate signals.
What to Do When AI Impressions Rise
- Protect the winning page’s facts, availability, author information and update date.
- Add clear internal links to the next useful product, collection or service page.
- Answer the immediate question early, then provide evidence, exceptions and practical detail.
- Keep product data consistent between the page, structured data and Merchant Center feed.
- Expand adjacent questions only when the business has something original to contribute.
If the page is visible but weak, run an AI search audit. Google’s own guidance remains refreshingly ordinary: make pages crawlable, indexable, useful, unique and easy to understand. There is no special AI schema or mandatory llms.txt shortcut.
Common Reporting Mistakes
Another common mistake is optimizing only the paragraph that looks quotable. AI experiences use Google’s broader Search systems. Thin “answer blocks” cannot compensate for duplicate collections, inaccessible JavaScript, weak product data or a site that has not built trust around the topic.
A Simple Monthly Reporting Template
- Executive summary: one sentence on the direction of AI visibility and one sentence on commercial impact.
- Visibility: total impressions, change, top countries and top devices.
- Page movement: five winners, five losers and page-type share.
- Outcome check: standard Search clicks plus organic revenue and assisted conversions for those URLs.
- Actions: three prioritized changes with an owner and review date.
Frequently Asked Questions
Does the report show AI Mode queries?
Not in the initial dedicated report. Google lists impressions, pages, countries, devices for Search and dates. Use ordinary Search Console query data for adjacent demand research, but do not present it as an exact list of AI prompts.
Are AI impressions included in total Search performance?
Yes. Google says the data remains part of the overall Performance report; the new report provides a separate view of generative-feature visibility.
Should we rewrite every page for AI?
No. Start with pages that already show material visibility or support valuable customer decisions. Improve clarity and evidence while keeping the page useful for the shopper who lands on it.